Project Summary Gastrointestinal stromal tumor (GIST) is a common form of soft tissue sarcoma, and our limited ability to predict the course of metastatic disease or the risk of recurrence following resection creates significant uncertainties in patient care. These critical unresolved problems underscore the deficit in biological understanding of factors involved in GIST oncogenesis and clinical behavior. We have previously characterized the enhancer and transcriptional landscape of GIST, using this information to identify a transcription factor (TF) network with elements predictive of patient outcomes. Within this network are core TF members present in all GIST subtypes and responsible for establishing basal GIST transcriptional output. Additional accessory TFs are present in a disease state-specific manner, being found exclusively in indolent or aggressive disease. Supporting the TF network, the MOZ histone acetyltransferase complex is uniquely responsible for establishing enhancers in GIST. Our central hypothesis is that core TFs and the MOZ chromatin regulatory complex generate the GIST transcriptional program, which is modified by state-specific accessory TFs to instruct disease behavior and determine clinical outcomes. Understanding how core TFs function to control the GIST epigenome, and how accessory TFs modulate this oncogenic framework, is of chief relevance to understanding the biology of this disease. Aim 1 of this proposal seeks to characterize how the core TF network members interact with enhancers and other transcriptional regulators to exert gene regulation. Because defined accessory TFs are expressed exclusively in a disease state-specific fashion, Aim 2 will determine how these factors modify the core TF network and influence transcriptional output through genetic disruption, and accessory TF expression in clinical samples will be used to associate their expression with clinical outcomes. Finally, Aim 3 will define the role of the MOZ complex, a unique dependency in GIST, in collaborating with TFs to generate the GIST transcriptional state. These studies will define the transcriptional machinery that underlies GIST, advance our understanding of molecular determinants of indolent and aggressive disease, and develop a biological framework for novel approaches to predicting clinical behavior that may impact the care of patients suffering from this disease. As a physician-scientist dedicated to understanding and treating sarcoma, my long-term goal is to develop an independent research program to generate a detailed understanding of GIST biology that will transform patient care. During my proposed training period, I will perform mentored research in the laboratory of Dr. Scott Armstrong at DFCI, with supportive co-mentorship from Dr. George Demetri; my outstanding advisory committee will further guide my research and career development. This mentorship, together with my exceptional institutional environment, access to superb educational programs, and myriad training opportunities, will enable me to advance my career goals and contribute to meaningful sarcoma research.